Investigation into implement status awareness for autonomous tractors

Green, Sam (2022) Investigation into implement status awareness for autonomous tractors. [USQ Project]

Text (Project)

Download (5MB) | Preview


Autonomous tractors are becoming more prevalent within agriculture. However, there is one thing that is hindering these machines from taking the role of an operator. The driver is currently the perception of the tractor and the only system that can stop the machine if something is about to or does go wrong. Developing a system that allows the tractor to have a better understanding of what it is doing and how its implement is operating will mean the requirement to have an operator sitting in the tractor seat is decreased.

The aim of the project is to determine if information exchanged via a tractor’s CAN Bus system can provide a sense of machine perception for autonomous applications.

The first part of the project involved designing a testing methodology that allowed a range of implements and operating conditions to be tested. From the testing methodology, a system was set up to collect messages from a range of given sensors on the tractor via the CAN Bus system. The testing was then carried out following the designed procedure while recording the operating status of the tractor to a computer. This raw data was then sent to John Deere to be converted from hexadecimal into usable Matlab file format. The data was then exported into Excel for further analysis and graphing.

From the results analysis, it can be concluded that the engine % load is the best indicator of an issue occurring with the implement. Rear hitch height, engine fuel usage and engine RPM are also parameters that showed promising results.

Overall, the sensitivity of the in-built sensors is not suitable without refinement to provide a good understanding of what the implements status is. This project has shown that CAN Bus based implement awareness systems have potential, however a large amount of refinement and work needs to be completed to ensure the system is consistent and reliable, and sufficiently responsive to allow an autonomous tractor to operate safely and efficiently.

Statistics for USQ ePrint 51904
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Faculty/School / Institute/Centre: Current – Faculty of Health, Engineering and Sciences - School of Agriculture and Environmental Science (1 Jan 2022 -)
Supervisors: Baillie, Justine
Qualification: Bachelor of Engineering (Honours) (Agriculture)
Date Deposited: 20 Jun 2023 05:06
Last Modified: 20 Jun 2023 05:06
Uncontrolled Keywords: Autonomous tractors; agriculture

Actions (login required)

View Item Archive Repository Staff Only